Persons with chronic kidney disease (CKD) are susceptible to an array of potentially harmful events related to their care (safety events). However, [there has been little work to establish a nomenclature for CKD-relevant safety events], to examine the prevalence of these safety events, or to determine the role of safety events in the incidence of poor CKD outcomes. Hypothesis: Patients with CKD have a high frequency of both general (PSI) and disease-specific patient safety incidents (CKD-PSI) that account for a significant proportion of adverse events leading to poor disease outcomes. Overall aims: Examine the prevalence of disease-specific CKD-PSI [designated as triggers] and general PSI in the CKD population, and determine whether the incidence of these PSIs cause adverse events which account for poor CKD outcomes. Specific aims: 1a) Determine the frequency of medication errors (one set of CKD-PSI) in a CKD cohort. 1b) Compare the rate of adverse events among subjects who have had one or more medication-related CKD-PSI as defined in (1a) to that of their counterparts who have not had such an event. 2a) Determine the incidence of electrolyte/metabolic disturbances including hyperkalemia and hypoglycemia (2nd set of CKD-PSIs) in a CKD cohort. 2b) Compare rate of adverse events in subjects who have had one or more electrolyte disturbance as defined in (2a) to that of their counterparts who have not had such an event. 3a) Examine the incidence of both AHRQ-defined PSI and a 3rd set of CKD-PSI (both ICD-9 code-derived) during hospitalizations of persons in CKD cohort. 3c) Compare the rate of adverse events in subjects who had a hospitalization and a PSI as defined in (3a) to that of their counterparts with a hospitalization and no such event. Study design: Retrospective (ancillary) study of an existing cohort. Study population: Participants (n = 3939) in the NIDDK- sponsored Chronic Renal Insufficiency Cohort (CRIC) study. Study Measurements: Medication records (30 day recall), including over-the-counter pills, determined at i) baseline, ii) the first semiannual interim phone visit, and iii) repeated annually. Medication errors will be determined using a renal drug reference cross-walk and CRIC estimates of glomerular filtration rate (GFR). Electrolyte disturbances (principally, but not exclusively, hyperkalemia and hypoglycemia) will be recorded from baseline and annual laboratory values. Additionally, ICD-9 codes from hospitalization over the course of CRIC observation will be reviewed for AHRQ- PSI and CKD-PSI. Outcomes: Participants will be tracked for adverse events including hospitalization, loss of renal function, ESRD and CVD events, and mortality. Analytic plans: Longitudinal and repeated measures methods will be employed to estimate the relationship between designated CKD-PSI (both as composite and relevant subgroups) and adverse events with focus on events occurring within one year of safety incidents. Relevance: This study will determine the extent to which harmful consequences of medical care (safety events) contribute to the characteristically poor outcomes of CKD.